| *** takedakn has joined #openstack-cinder | 00:00 | |
| *** chlong has quit IRC | 00:01 | |
| *** chlong has joined #openstack-cinder | 00:02 | |
| *** takedakn has quit IRC | 00:03 | |
| *** takedakn has joined #openstack-cinder | 00:03 | |
| *** fischerw has quit IRC | 00:04 | |
| *** chlong has quit IRC | 00:06 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Implement clone_image caching on SolidFire https://review.openstack.org/142859 | 00:18 |
|---|---|---|
| *** jaypipes has quit IRC | 00:18 | |
| *** jungleboyj has joined #openstack-cinder | 00:18 | |
| *** Masahiro has joined #openstack-cinder | 00:20 | |
| thingee | jgriffith: hey so I stated looking at some of the stub issues from the unit tests that are failing with cinder objects. | 00:22 |
| *** scottda_ has quit IRC | 00:23 | |
| thingee | jgriffith: I noticed the first set of tests failing are using the stubs from cinder.tests.api.v2.stubs | 00:23 |
| thingee | which is somewhat good, because it's using shared fixtures | 00:23 |
| jungleboyj | thingee: jgriffith smcginnis hemna Here is the fix for the _lazy in stable/juno: https://review.openstack.org/145642 | 00:23 |
| thingee | just they need to be improved | 00:23 |
| *** Masahiro has quit IRC | 00:24 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Fix drbd driver to load without 3'rd party libs https://review.openstack.org/145644 | 00:25 |
| *** IanGovett has quit IRC | 00:27 | |
| thingee | jungleboyj: thanks, lgtm once jenkins signs off | 00:28 |
| *** chlong has joined #openstack-cinder | 00:29 | |
| jungleboyj | thingee: Thank you! | 00:31 |
| thingee | jungleboyj: https://review.openstack.org/#/c/145613/ | 00:31 |
| thingee | jungleboyj: http://paste.openstack.org/raw/155906/ | 00:32 |
| vishy | thingee, jgriffith so it turns out the allocated space code is pretty broken | 00:32 |
| vishy | allocated_capacity_gb for scheduling | 00:32 |
| jungleboyj | I saw you pushed that up thingee Let me look here. | 00:33 |
| vishy | it is tracked in the manager but if volume creation fails in the driver it doesn’t add to the allocated_capacity | 00:33 |
| vishy | but it deletes it when you delete the volume | 00:33 |
| *** smoriya has joined #openstack-cinder | 00:34 | |
| *** takedakn has quit IRC | 00:34 | |
| jungleboyj | thingee: Hmmm, that is pretty simple. | 00:35 |
| thingee | jungleboyj: suspicious | 00:36 |
| jungleboyj | thingee: suspicious? | 00:36 |
| thingee | yeah simple code, probably something is wrong | 00:36 |
| thingee | my manual testing seems to disagree, but wanted other eyes on it | 00:36 |
| jungleboyj | thingee: No, not at all. | 00:36 |
| *** annashen has joined #openstack-cinder | 00:37 | |
| *** bswartz has joined #openstack-cinder | 00:37 | |
| thingee | vishy: interesting | 00:37 |
| jungleboyj | My eyes on it are probably not the best as the client and session code are all a mystery to me. | 00:37 |
| jungleboyj | My only concern is what if _get_project returns None. What heppsn then? | 00:38 |
| *** Mandell has quit IRC | 00:38 | |
| thingee | jungleboyj: so I was talking with jamielennox about that. My understanding was that happens if a keystone plugin doesn't implement that. But apparently most are based off identity. | 00:39 |
| thingee | so it shouldn't be a problem, but a valid concern | 00:39 |
| thingee | most = all? | 00:39 |
| thingee | I dunno. | 00:39 |
| * jungleboyj really needs to figure out the keystone stuff. | 00:39 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Deal with tgt already exists errors https://review.openstack.org/145608 | 00:40 |
| jgriffith | vishy: son of a.... | 00:41 |
| jgriffith | vishy: so it's broken for a number of other reasons, but that one isn't one that I had on the list | 00:41 |
| vishy | jgriffith: also in the icehouse code it doesn’t count available volumes to the total | 00:42 |
| jamielennox | thingee: is that running, you have def _get_project(self) and self._get_project_id() | 00:42 |
| vishy | but that has been fixed | 00:42 |
| thingee | jamielennox: ha, apparently. | 00:43 |
| jgriffith | vishy: I'll spend some time going through tomorrow and logging bugs/fixing as I go | 00:43 |
| thingee | jamielennox: I don't know how | 00:43 |
| jgriffith | vishy: unless you're already doing that :) | 00:43 |
| vishy | jgriffith: thing is i’m not quite sure how to fix it | 00:43 |
| jamielennox | my thought would be if there is no project_id returned then just don't try and append anything | 00:43 |
| jgriffith | vishy: might not be able to, that's what we've been fighting about for a year now | 00:43 |
| *** km has quit IRC | 00:43 | |
| vishy | the problem is if the driver errors is the space used or not? | 00:44 |
| vishy | it seems like the driver should be reporting this data | 00:44 |
| jgriffith | vishy: yeah, and if we fix the stats reporting they will | 00:44 |
| jgriffith | vishy: so what I'm thinking is just brute force the change rather than waiting for everybody to agree | 00:44 |
| jgriffith | vishy: in other words, go purely from stats reported back to mgr from driver | 00:45 |
| vishy | jgriffith: i was thinking you could gust get it from the driver but default it to total - free if it isn’t set | 00:45 |
| jgriffith | vishy: we already rely on that soley for scheduler (or should) | 00:45 |
| jamielennox | thingee: what i'd love is that for every path in v2 with a URL containing the project_id you validate that project_id in URL == project_id in token then redirect the call to the same url without the project_id | 00:45 |
| vishy | but that doesn’t work for those drivers that are passing back total as ‘unknown' | 00:45 |
| jamielennox | thingee: a path to removing the project_id from cinder calls altogether | 00:45 |
| jamielennox | s/calls/urls | 00:45 |
| vishy | maybe it is just ‘unknown’ for those drivers? | 00:45 |
| *** scottda_ has joined #openstack-cinder | 00:46 | |
| thingee | jamielennox: k | 00:46 |
| jgriffith | vishy: so maybe that's a good motivation actually. | 00:46 |
| jamielennox | big job, small pay off short term | 00:47 |
| jgriffith | vishy: if you report unknown or inf we'll use the crappy internal guessing | 00:47 |
| jgriffith | so it's in your best interest to fix it | 00:47 |
| jgriffith | vishy: you around tomorrow? | 00:47 |
| jungleboyj | thingee: The paste link you have doesn't work. | 00:51 |
| *** scottda_ has quit IRC | 00:52 | |
| thingee | yeah I think there might be something wrong with paste...it was working earlier | 00:52 |
| jungleboyj | Ok. Weird. | 00:53 |
| vishy | jgriffith: probably | 00:55 |
| *** Mandell has joined #openstack-cinder | 00:56 | |
| *** annashen has quit IRC | 00:58 | |
| *** dank has quit IRC | 01:04 | |
| *** alexpilotti has quit IRC | 01:08 | |
| jungleboyj | jgriffith: Good point on the _lazy review. I need to drop off agian but I will fix that up a little later and push another patch. | 01:08 |
| *** jdurgin has quit IRC | 01:09 | |
| openstackgerrit | wanghao proposed openstack/cinder: Implement function of import/export snapshots https://review.openstack.org/144590 | 01:15 |
| *** jungleboyj has quit IRC | 01:15 | |
| *** scottda_ has joined #openstack-cinder | 01:18 | |
| *** ebalduf has quit IRC | 01:19 | |
| openstackgerrit | Mitsuhiro Tanino proposed openstack/cinder: Remove unused variables from ensure_export() https://review.openstack.org/145381 | 01:20 |
| *** scottda_ has quit IRC | 01:20 | |
| *** tsekiyam_ has joined #openstack-cinder | 01:23 | |
| mudassirlatif | hey jgriffith, can you please approve this (again) https://review.openstack.org/#/c/145386/ …. I had approval from you and workflow +1 from someone else but had to rebase because of the oslo _lazy bug. It’s a easy to examine change regarding i18n fixes | 01:24 |
| *** tsekiyama has quit IRC | 01:26 | |
| openstackgerrit | Anthony Lee proposed openstack/cinder: Fixed server name being retained after detach in LeftHand https://review.openstack.org/124145 | 01:27 |
| openstackgerrit | Anthony Lee proposed openstack/cinder: Remove locks from LeftHand driver https://review.openstack.org/141475 | 01:28 |
| *** tsekiyam_ has quit IRC | 01:28 | |
| *** mtanino has quit IRC | 01:28 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Use cinder.utils.execute directly https://review.openstack.org/145630 | 01:29 |
| *** zhithuang has joined #openstack-cinder | 01:31 | |
| *** zhithuang is now known as winston-d_ | 01:31 | |
| *** leeantho has quit IRC | 01:33 | |
| thingee | winston-d_: ping | 01:41 |
| *** nosnos has joined #openstack-cinder | 01:54 | |
| *** thingee has quit IRC | 01:57 | |
| *** Mandell has quit IRC | 02:02 | |
| *** elazro has quit IRC | 02:06 | |
| *** elazro has joined #openstack-cinder | 02:06 | |
| *** ryanpetrello has quit IRC | 02:08 | |
| *** haomaiwang has joined #openstack-cinder | 02:08 | |
| *** Masahiro has joined #openstack-cinder | 02:09 | |
| *** _cjones_ has quit IRC | 02:11 | |
| *** chlong has quit IRC | 02:11 | |
| *** ebalduf has joined #openstack-cinder | 02:14 | |
| *** winston-d_ has quit IRC | 02:15 | |
| *** winston-d_ has joined #openstack-cinder | 02:15 | |
| *** julim has joined #openstack-cinder | 02:16 | |
| *** ebalduf has quit IRC | 02:21 | |
| *** ebalduf has joined #openstack-cinder | 02:24 | |
| *** Mandell has joined #openstack-cinder | 02:27 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/145661 | 02:28 |
| *** ebalduf has quit IRC | 02:28 | |
| *** winston-d_ has quit IRC | 02:31 | |
| *** thingee has joined #openstack-cinder | 02:34 | |
| openstackgerrit | Mudassir Latif proposed openstack/cinder: Add replication and PG setup methods on PureISCSIDriver https://review.openstack.org/145090 | 02:36 |
| *** ling-yun-1 has joined #openstack-cinder | 02:37 | |
| *** zengyunling__ has joined #openstack-cinder | 02:39 | |
| *** winston-d_ has joined #openstack-cinder | 02:39 | |
| *** ling-yun has quit IRC | 02:40 | |
| *** kaisers has joined #openstack-cinder | 02:41 | |
| *** fandi has quit IRC | 02:42 | |
| *** ling-yun-1 has quit IRC | 02:42 | |
| thingee | winston-d_: ping | 02:42 |
| *** mberlin1 has joined #openstack-cinder | 02:43 | |
| *** mberlin has quit IRC | 02:43 | |
| *** kaisers1 has quit IRC | 02:44 | |
| openstackgerrit | Thang Pham proposed openstack/cinder: WIP: Cinder objects base https://review.openstack.org/131873 | 02:44 |
| openstackgerrit | Thang Pham proposed openstack/cinder: WIP: Snapshot object https://review.openstack.org/133566 | 02:45 |
| *** winston-d_ has quit IRC | 02:46 | |
| *** Mandell has quit IRC | 02:48 | |
| *** winston-d_ has joined #openstack-cinder | 02:52 | |
| winston-d_ | thingee: pong | 02:53 |
| *** ryanpetrello has joined #openstack-cinder | 02:53 | |
| winston-d_ | thingee: sorry i am having a bad internet connection today. | 02:53 |
| *** annashen has joined #openstack-cinder | 02:56 | |
| thingee | winston-d_: heh seems like a trend today :) | 03:00 |
| thingee | so not sure if you saw vishy's earlier comments | 03:00 |
| thingee | in #openstack-cinder | 03:00 |
| winston-d_ | thingee: I didn't. what is it? | 03:02 |
| thingee | winston-d_: allocated_capacity_gb code for the scheduler appears to be broken. Since it's tracked in the manager, if the volume creation fails in the driver, it doesn't add the allocated_capacity. | 03:02 |
| *** Mandell has joined #openstack-cinder | 03:02 | |
| winston-d_ | thingee: is there a bug? | 03:02 |
| thingee | winston-d_: convo is here http://eavesdrop.openstack.org/irclogs/%23openstack-cinder/%23openstack-cinder.2015-01-08.log | 03:03 |
| *** lcurtis has joined #openstack-cinder | 03:03 | |
| thingee | winston-d_: search for timestamp '2015-01-08T00:32:24' | 03:03 |
| winston-d_ | thingee: can't believe that I didn't know that we have log in cinder channel now! | 03:04 |
| winston-d_ | thingee: looking, thx. | 03:04 |
| thingee | winston-d_: oh yeah, pretty recently turned on. | 03:05 |
| *** winston-d_ has quit IRC | 03:11 | |
| *** winston-d_ has joined #openstack-cinder | 03:13 | |
| *** Mandell has quit IRC | 03:16 | |
| *** scottda_ has joined #openstack-cinder | 03:18 | |
| *** mudassirlatif has quit IRC | 03:20 | |
| *** patrickeast has quit IRC | 03:20 | |
| openstackgerrit | Alan Jiang proposed openstack/cinder: rbd calls block eventlet threads https://review.openstack.org/145678 | 03:21 |
| *** bill_az has quit IRC | 03:22 | |
| *** ebalduf has joined #openstack-cinder | 03:25 | |
| *** Mandell has joined #openstack-cinder | 03:27 | |
| *** scottda_ has quit IRC | 03:29 | |
| *** ebalduf has quit IRC | 03:30 | |
| *** scottda_ has joined #openstack-cinder | 03:36 | |
| *** nosnos has quit IRC | 03:43 | |
| *** ryanpetrello has quit IRC | 03:49 | |
| *** _cjones_ has joined #openstack-cinder | 03:54 | |
| *** _cjones_ has joined #openstack-cinder | 03:54 | |
| *** ebalduf has joined #openstack-cinder | 03:55 | |
| *** annashen has quit IRC | 03:55 | |
| *** chlong has joined #openstack-cinder | 03:57 | |
| *** k_ has joined #openstack-cinder | 03:59 | |
| *** bkopilov has quit IRC | 04:01 | |
| *** chlong has quit IRC | 04:02 | |
| *** jungleboyj has joined #openstack-cinder | 04:09 | |
| openstackgerrit | John Griffith proposed openstack/cinder: Use cinder.utils.execute directly https://review.openstack.org/145630 | 04:10 |
| thingee | jgriffith: you around? | 04:17 |
| *** chlong has joined #openstack-cinder | 04:18 | |
| *** _cjones_ has quit IRC | 04:20 | |
| *** fandi has joined #openstack-cinder | 04:22 | |
| *** fandi has quit IRC | 04:23 | |
| *** fandi has joined #openstack-cinder | 04:24 | |
| *** fandi has quit IRC | 04:24 | |
| *** fandi has joined #openstack-cinder | 04:26 | |
| *** annashen has joined #openstack-cinder | 04:28 | |
| *** scottda_ has quit IRC | 04:30 | |
| openstackgerrit | Mike Perez proposed openstack/python-cinderclient: Remove unused service catalog module https://review.openstack.org/145686 | 04:35 |
| *** nosnos has joined #openstack-cinder | 04:36 | |
| *** ebalduf has quit IRC | 04:39 | |
| *** lcurtis has quit IRC | 04:39 | |
| *** annashen has quit IRC | 04:42 | |
| *** ebalduf has joined #openstack-cinder | 04:42 | |
| *** rakesh_ has joined #openstack-cinder | 04:45 | |
| *** rakesh_ is now known as Guest45532 | 04:45 | |
| *** Longgeek has joined #openstack-cinder | 04:45 | |
| Guest45532 | rakesh_mishra510 | 04:46 |
| *** Guest45532 is now known as rakesh_mishra | 04:46 | |
| *** ebalduf has quit IRC | 04:47 | |
| *** david-lyle has quit IRC | 04:50 | |
| *** winston-d_ has quit IRC | 04:51 | |
| *** dannywilson has quit IRC | 04:51 | |
| *** annashen has joined #openstack-cinder | 05:00 | |
| *** winston-d_ has joined #openstack-cinder | 05:05 | |
| *** anshul has joined #openstack-cinder | 05:06 | |
| *** annashen has quit IRC | 05:10 | |
| *** _cjones_ has joined #openstack-cinder | 05:12 | |
| *** winston-d_ has quit IRC | 05:14 | |
| *** winston-d_ has joined #openstack-cinder | 05:14 | |
| *** _cjones_ has quit IRC | 05:17 | |
| *** annashen has joined #openstack-cinder | 05:19 | |
| *** annashen has quit IRC | 05:19 | |
| openstackgerrit | Alan Jiang proposed openstack/cinder: rbd calls block eventlet threads https://review.openstack.org/145678 | 05:26 |
| *** annashen has joined #openstack-cinder | 05:26 | |
| *** BharatK has joined #openstack-cinder | 05:27 | |
| *** bill_az has joined #openstack-cinder | 05:30 | |
| *** annashen has quit IRC | 05:30 | |
| *** Mandell has joined #openstack-cinder | 05:32 | |
| *** nshaikh has joined #openstack-cinder | 05:36 | |
| *** ebalduf has joined #openstack-cinder | 05:43 | |
| openstackgerrit | Mike Perez proposed openstack/python-cinderclient: Remove v1 support https://review.openstack.org/145688 | 05:48 |
| *** ebalduf has quit IRC | 05:48 | |
| *** rakesh_mishra has quit IRC | 05:52 | |
| *** winston-d_ has quit IRC | 05:53 | |
| *** rakesh_mishra has joined #openstack-cinder | 05:58 | |
| *** mudassirlatif has joined #openstack-cinder | 05:58 | |
| *** winston-d_ has joined #openstack-cinder | 06:00 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Imported Translations from Transifex https://review.openstack.org/144933 | 06:04 |
| *** rushil has quit IRC | 06:08 | |
| *** RJ_ has joined #openstack-cinder | 06:12 | |
| *** mudassirlatif has quit IRC | 06:14 | |
| *** mudassirlatif has joined #openstack-cinder | 06:17 | |
| openstackgerrit | Trung Trinh proposed openstack/cinder: Delete volume attached to already-deleted VM https://review.openstack.org/142006 | 06:19 |
| *** mudassirlatif has quit IRC | 06:26 | |
| *** winston-d_ has quit IRC | 06:27 | |
| *** winston-d_ has joined #openstack-cinder | 06:29 | |
| *** jamielennox is now known as jamielennox|away | 06:33 | |
| *** bkopilov has joined #openstack-cinder | 06:33 | |
| *** nshaikh has quit IRC | 06:41 | |
| *** ebalduf has joined #openstack-cinder | 06:44 | |
| anish | jgriffith: you had mentioned previously that the generic iscsi-helper code was being re-written for kilo | 06:46 |
| anish | is there a place to follow this so that I make sure I don't end up writing deprecated code wrt https://review.openstack.org/#/c/134398/ | 06:47 |
| *** ebalduf has quit IRC | 06:50 | |
| *** pradipta_away is now known as pradipta | 06:52 | |
| *** nlevinki has joined #openstack-cinder | 06:53 | |
| *** mogami has joined #openstack-cinder | 06:54 | |
| thingee | anish: targets are now being created in https://github.com/openstack/cinder/tree/master/cinder/volume/targets | 06:56 |
| thingee | example, tgt and lio being iscsi helpers | 06:56 |
| *** nkrinner has joined #openstack-cinder | 06:57 | |
| anish | yep, saw that | 06:57 |
| anish | iet seems to have gone away ? | 06:57 |
| *** mogami has quit IRC | 06:59 | |
| openstackgerrit | Mike Perez proposed openstack/cinder: Remove v1 server support https://review.openstack.org/145703 | 07:00 |
| thingee | anish: not sure about iet, sorry | 07:01 |
| anish | no worries, I just ask because I was working on this before the holidays, and a git pull basically broke everything | 07:02 |
| anish | thingee: just wanted to confirm if there are more changes coming | 07:02 |
| *** Longgeek has quit IRC | 07:02 | |
| *** lpetrut has joined #openstack-cinder | 07:03 | |
| thingee | anish: We have bug fixes, but that should be it in terms of big move changes | 07:03 |
| anish | excellent ! | 07:03 |
| *** Longgeek has joined #openstack-cinder | 07:03 | |
| *** afazekas is now known as __afazekas | 07:04 | |
| openstackgerrit | Trung Trinh proposed openstack/cinder: Delete volume attached to already-deleted VM https://review.openstack.org/142006 | 07:04 |
| *** winston-d_ has quit IRC | 07:04 | |
| *** winston-d_ has joined #openstack-cinder | 07:06 | |
| *** coolsvap|afk is now known as coolsvap | 07:08 | |
| *** e0ne has joined #openstack-cinder | 07:11 | |
| *** karimb has joined #openstack-cinder | 07:11 | |
| *** bill_az has quit IRC | 07:13 | |
| *** EduardBiceriMate has joined #openstack-cinder | 07:17 | |
| *** Mandell has quit IRC | 07:23 | |
| flip214 | jgriffith: now I am ... but that won't help much, I guess. | 07:24 |
| *** winston-d_ has quit IRC | 07:24 | |
| *** Mandell has joined #openstack-cinder | 07:25 | |
| thingee | flip214: :) | 07:25 |
| flip214 | thingee: I guess I know what john wants, already reviewing / replying. | 07:28 |
| flip214 | thingee: [x] 145644 done | 07:31 |
| *** k4n0 has joined #openstack-cinder | 07:38 | |
| *** k4n0 has quit IRC | 07:42 | |
| openstackgerrit | Marc Koderer proposed openstack/cinder: Update README.rst to current state https://review.openstack.org/145519 | 07:45 |
| *** afazekas_ has joined #openstack-cinder | 07:46 | |
| *** chlong has quit IRC | 07:47 | |
| *** TobiasE has joined #openstack-cinder | 07:47 | |
| *** nshaikh has joined #openstack-cinder | 07:51 | |
| *** dannywilson has joined #openstack-cinder | 07:52 | |
| *** e0ne has quit IRC | 07:54 | |
| *** dannywilson has quit IRC | 07:57 | |
| *** rakesh_mishra has quit IRC | 07:59 | |
| *** rakesh_mishra has joined #openstack-cinder | 08:00 | |
| *** thingee has quit IRC | 08:17 | |
| *** sgotliv has joined #openstack-cinder | 08:18 | |
| *** delattec has joined #openstack-cinder | 08:21 | |
| *** rakesh_mishra has quit IRC | 08:21 | |
| *** cdelatte has quit IRC | 08:23 | |
| openstackgerrit | TaoBai proposed openstack/cinder: IBM Storwize driver Consistency Group Implementation https://review.openstack.org/142683 | 08:25 |
| *** ndipanov has joined #openstack-cinder | 08:29 | |
| *** pradipta is now known as pradipta_away | 08:31 | |
| *** alexpilotti has joined #openstack-cinder | 08:31 | |
| *** alexpilotti has quit IRC | 08:36 | |
| *** alexpilotti has joined #openstack-cinder | 08:36 | |
| *** Mandell has quit IRC | 08:38 | |
| *** jkraj has joined #openstack-cinder | 08:40 | |
| *** ebalduf has joined #openstack-cinder | 08:48 | |
| *** anshul has quit IRC | 08:51 | |
| *** anshul has joined #openstack-cinder | 08:53 | |
| *** ebalduf has quit IRC | 08:53 | |
| *** jpich has joined #openstack-cinder | 08:54 | |
| *** jistr has joined #openstack-cinder | 09:10 | |
| *** jordanP has joined #openstack-cinder | 09:12 | |
| openstackgerrit | Jeegn Chen proposed openstack/cinder: EMC VNX Cinder Driver Update https://review.openstack.org/141629 | 09:13 |
| *** lpetrut has quit IRC | 09:17 | |
| *** dalgaaf has joined #openstack-cinder | 09:26 | |
| *** tshefi has joined #openstack-cinder | 09:28 | |
| *** alecv has joined #openstack-cinder | 09:31 | |
| *** Guest57286 has joined #openstack-cinder | 09:38 | |
| *** nellysmitt has joined #openstack-cinder | 09:45 | |
| *** jordanP has quit IRC | 09:45 | |
| *** jordanP has joined #openstack-cinder | 09:45 | |
| *** ebalduf has joined #openstack-cinder | 09:49 | |
| *** RJ_ has left #openstack-cinder | 09:51 | |
| *** sgotliv has quit IRC | 09:52 | |
| *** ebalduf has quit IRC | 09:54 | |
| *** dulek has joined #openstack-cinder | 09:55 | |
| *** sgotliv has joined #openstack-cinder | 10:06 | |
| *** lpetrut has joined #openstack-cinder | 10:09 | |
| *** rakesh__mishra has joined #openstack-cinder | 10:10 | |
| *** TobiasE has quit IRC | 10:17 | |
| *** RJ_ has joined #openstack-cinder | 10:17 | |
| *** e0ne has joined #openstack-cinder | 10:23 | |
| *** Masahiro has quit IRC | 10:27 | |
| *** rushiagr_away is now known as rushiagr | 10:32 | |
| *** rushiagr is now known as rushiagr_away | 10:33 | |
| *** smoriya has quit IRC | 10:42 | |
| *** e0ne_ has joined #openstack-cinder | 10:48 | |
| *** e0ne has quit IRC | 10:48 | |
| *** ebalduf has joined #openstack-cinder | 10:50 | |
| *** ebalduf has quit IRC | 10:55 | |
| erlon_ | . | 10:55 |
| erlon_ | thingee: ping | 10:55 |
| *** haomaiwang has quit IRC | 10:56 | |
| *** karimb has quit IRC | 10:57 | |
| RJ_ | hi | 11:05 |
| RJ_ | i am facing the error while creating the volume with type having capabilities:replication= True in key | 11:06 |
| RJ_ | ..? | 11:06 |
| RJ_ | http://paste.openstack.org/show/155969/ | 11:07 |
| *** deepakcs has joined #openstack-cinder | 11:10 | |
| *** TobiasE has joined #openstack-cinder | 11:25 | |
| *** chlong has joined #openstack-cinder | 11:43 | |
| *** ebalduf has joined #openstack-cinder | 11:51 | |
| *** BharatK has quit IRC | 11:54 | |
| *** ebalduf has quit IRC | 11:55 | |
| *** Miouge has quit IRC | 12:02 | |
| *** Miouge has joined #openstack-cinder | 12:03 | |
| *** ho has quit IRC | 12:08 | |
| *** Miouge has quit IRC | 12:08 | |
| *** Masahiro has joined #openstack-cinder | 12:16 | |
| *** Masahiro has quit IRC | 12:20 | |
| openstackgerrit | git-harry proposed openstack/python-cinderclient: v2 error message grammatical error https://review.openstack.org/145767 | 12:24 |
| *** sandywalsh has quit IRC | 12:24 | |
| *** k_ has quit IRC | 12:25 | |
| *** rakesh__mishra has quit IRC | 12:26 | |
| *** nosnos has quit IRC | 12:30 | |
| *** deepakcs has quit IRC | 12:35 | |
| *** karimb has joined #openstack-cinder | 12:38 | |
| *** IanGovett has joined #openstack-cinder | 12:40 | |
| *** dulek has quit IRC | 12:50 | |
| *** ebalduf has joined #openstack-cinder | 12:51 | |
| *** dulek has joined #openstack-cinder | 12:53 | |
| *** ryanpetrello has joined #openstack-cinder | 12:54 | |
| *** ebalduf has quit IRC | 12:56 | |
| *** sgotliv_ has joined #openstack-cinder | 12:59 | |
| *** sgotliv has quit IRC | 13:00 | |
| *** Miouge has joined #openstack-cinder | 13:03 | |
| openstackgerrit | Pratik Das proposed openstack/python-cinderclient: Removed --name from cinder list as there was no NAME column in cinder volumes database. https://review.openstack.org/145774 | 13:07 |
| *** anshul has quit IRC | 13:13 | |
| *** ryanpetrello has quit IRC | 13:16 | |
| *** ryanpetrello has joined #openstack-cinder | 13:17 | |
| *** bswartz has quit IRC | 13:19 | |
| openstackgerrit | nikeshmahalka proposed openstack/cinder: Add iSCSI SCST Target support to cinder. https://review.openstack.org/145778 | 13:24 |
| *** alexpilotti has quit IRC | 13:26 | |
| *** bill_az has joined #openstack-cinder | 13:29 | |
| openstackgerrit | Anton Arefiev proposed openstack/cinder: Import only modules in volume manager https://review.openstack.org/145780 | 13:33 |
| *** nikesh_vedams has joined #openstack-cinder | 13:39 | |
| *** ryanpetrello has quit IRC | 13:50 | |
| *** e0ne_ has quit IRC | 13:51 | |
| *** kaufer has joined #openstack-cinder | 13:52 | |
| *** ebalduf has joined #openstack-cinder | 13:52 | |
| *** jcru has joined #openstack-cinder | 13:54 | |
| *** timcl has joined #openstack-cinder | 13:55 | |
| *** bswartz has joined #openstack-cinder | 13:56 | |
| *** jungleboyj has quit IRC | 13:56 | |
| *** ebalduf has quit IRC | 13:56 | |
| *** jaypipes has joined #openstack-cinder | 13:58 | |
| *** e0ne has joined #openstack-cinder | 13:58 | |
| *** nlevinki has quit IRC | 14:01 | |
| nikesh_vedams | Hi,i have a small doubt,i want to create a Gerrit CI account and i already have a launchpad account for submitting patch,so for CI account can i use this launchpad account? | 14:03 |
| nikesh_vedams | i used this launchpad account for registering blueprint and submitting patch for blueprint | 14:04 |
| larsks | nikesh_vedams: yes, you can use that same account with openstack gerrit. | 14:04 |
| *** Masahiro has joined #openstack-cinder | 14:05 | |
| nikesh_vedams | larsks: this link is confusing for me http://ci.openstack.org/third_party.html | 14:06 |
| nikesh_vedams | larsks: under Creating a Service Account | 14:07 |
| nikesh_vedams | larsks: its saying "You may already have an existing personal account in Launchpad, but you should create a new and entirely separate account for this purpose." | 14:07 |
| larsks | nikesh_vedams: that page is about setting up a third party CI envrionment, not about submitting reviews to gerrit. Which are you trying to do? | 14:08 |
| larsks | I may have misunderstood. | 14:08 |
| nikesh_vedams | larsks: i want to set up a CI setup | 14:08 |
| larsks | Okay. In that case, you *would* want to create a new account specifically for the CI setup. | 14:09 |
| *** Yogi1 has joined #openstack-cinder | 14:09 | |
| larsks | So that the results of CI testing can be distinguished from your own comments on patches. | 14:09 |
| *** akerr has joined #openstack-cinder | 14:09 | |
| *** Masahiro has quit IRC | 14:09 | |
| nikesh_vedams | larsks: ok got it,thanks | 14:11 |
| nikesh_vedams | larsks:i want to set up "sos-ci",but stuggling where to start | 14:12 |
| nikesh_vedams | larsks:https://github.com/j-griffith/sos-ci | 14:12 |
| *** esker has joined #openstack-cinder | 14:12 | |
| *** esker has quit IRC | 14:13 | |
| larsks | nikesh_vedams: That I'm not familiar with. | 14:14 |
| nikesh_vedams | larsks:ok no probs :) | 14:15 |
| *** nlevinki has joined #openstack-cinder | 14:16 | |
| *** ndipanov has quit IRC | 14:19 | |
| *** ndipanov has joined #openstack-cinder | 14:21 | |
| *** mriedem has joined #openstack-cinder | 14:21 | |
| openstackgerrit | Boris Pavlovic proposed openstack/cinder: Add more rally benchmarks related to Cinder https://review.openstack.org/145595 | 14:25 |
| openstackgerrit | Boris Pavlovic proposed openstack/cinder: Add more rally benchmarks related to Cinder https://review.openstack.org/145595 | 14:28 |
| *** xyang has joined #openstack-cinder | 14:30 | |
| *** alexpilotti has joined #openstack-cinder | 14:30 | |
| openstackgerrit | Steven Kaufer proposed openstack/cinder: GET volumes API sorting REST/volume/DB updates https://review.openstack.org/141915 | 14:31 |
| openstackgerrit | Steven Kaufer proposed openstack/cinder: GET volumes API sorting enhancements common utilities https://review.openstack.org/141914 | 14:31 |
| *** nshaikh has quit IRC | 14:33 | |
| *** dustins has joined #openstack-cinder | 14:35 | |
| *** thangp has joined #openstack-cinder | 14:42 | |
| *** ebalduf has joined #openstack-cinder | 14:53 | |
| *** lcurtis has joined #openstack-cinder | 14:53 | |
| *** ebalduf has quit IRC | 14:58 | |
| *** ling-yun-1 has joined #openstack-cinder | 14:58 | |
| *** zengyunling__ has quit IRC | 15:02 | |
| *** ndipanov has quit IRC | 15:05 | |
| *** Miouge has quit IRC | 15:09 | |
| *** EduardBiceriMate has quit IRC | 15:09 | |
| *** mtanino has joined #openstack-cinder | 15:11 | |
| *** ryanpetrello has joined #openstack-cinder | 15:11 | |
| *** rushil has joined #openstack-cinder | 15:12 | |
| *** akerr_ has joined #openstack-cinder | 15:13 | |
| *** lpabon has joined #openstack-cinder | 15:14 | |
| *** akerr has quit IRC | 15:14 | |
| *** vmtyler has joined #openstack-cinder | 15:15 | |
| *** ebalduf has joined #openstack-cinder | 15:17 | |
| *** nlevinki has quit IRC | 15:17 | |
| *** nlevinki has joined #openstack-cinder | 15:18 | |
| *** ndipanov has joined #openstack-cinder | 15:18 | |
| *** Longgeek has quit IRC | 15:19 | |
| *** Longgeek has joined #openstack-cinder | 15:21 | |
| *** thingee has joined #openstack-cinder | 15:24 | |
| *** tbarron has joined #openstack-cinder | 15:25 | |
| *** takedakn has joined #openstack-cinder | 15:25 | |
| *** takedakn has quit IRC | 15:27 | |
| *** takedakn has joined #openstack-cinder | 15:27 | |
| *** takedakn has quit IRC | 15:28 | |
| *** takedakn has joined #openstack-cinder | 15:28 | |
| *** Mandell has joined #openstack-cinder | 15:32 | |
| openstackgerrit | Mudassir Latif proposed openstack/cinder: Add replication and PG setup methods on PureISCSIDriver https://review.openstack.org/145090 | 15:32 |
| *** Miouge has joined #openstack-cinder | 15:33 | |
| *** tshefi has quit IRC | 15:35 | |
| *** takedakn has quit IRC | 15:35 | |
| *** jecarey has joined #openstack-cinder | 15:36 | |
| *** alecv has quit IRC | 15:38 | |
| *** vmtyler has quit IRC | 15:39 | |
| *** fischerw has joined #openstack-cinder | 15:42 | |
| *** alecv has joined #openstack-cinder | 15:43 | |
| *** Mandell has quit IRC | 15:43 | |
| *** e0ne is now known as e0ne_ | 15:47 | |
| *** lpetrut has quit IRC | 15:50 | |
| *** nlevinki has quit IRC | 15:51 | |
| *** Masahiro has joined #openstack-cinder | 15:53 | |
| *** mberlin1 has quit IRC | 15:54 | |
| *** mberlin has joined #openstack-cinder | 15:55 | |
| *** coolsvap is now known as coolsvap|afk | 15:55 | |
| *** e0ne_ has quit IRC | 15:57 | |
| *** TobiasE has quit IRC | 15:58 | |
| *** tsekiyama has joined #openstack-cinder | 15:58 | |
| *** Masahiro has quit IRC | 15:58 | |
| *** TobiasE has joined #openstack-cinder | 15:59 | |
| *** dulek has quit IRC | 16:00 | |
| *** dulek has joined #openstack-cinder | 16:00 | |
| *** RJ_ has left #openstack-cinder | 16:03 | |
| *** changbl has quit IRC | 16:04 | |
| annegentle_ | hey thingee, I could use help triaging https://bugs.launchpad.net/openstack-api-site/+bug/1331246 -- basically what are attachments and snapshot_id in responses for GET volume for v1 API? | 16:08 |
| *** alecv has quit IRC | 16:09 | |
| *** e0ne has joined #openstack-cinder | 16:09 | |
| *** changbl has joined #openstack-cinder | 16:15 | |
| *** nshaikh has joined #openstack-cinder | 16:16 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: PEP8(E265): block comment should start with '# ' https://review.openstack.org/145821 | 16:16 |
| *** afazekas_ has quit IRC | 16:18 | |
| *** dulek has quit IRC | 16:19 | |
| *** lpabon_ has joined #openstack-cinder | 16:20 | |
| openstackgerrit | Rushil Chugh proposed openstack/cinder: Garbage Remains when Attached Volume is Migrated with NFS Driver https://review.openstack.org/143431 | 16:20 |
| *** xyang has quit IRC | 16:23 | |
| *** alecv has joined #openstack-cinder | 16:23 | |
| e0ne | jgriffith, eharney: hi! some thoughts about https://bugs.launchpad.net/cinder/+bug/1373513: i've fixed similar issue in Ubuntu upgrading lvm2 package to v2.0.111. for now, i've got errors on centos with lvm2 v2.0.100. may be it isn't cinder issue. i'm still investigating it | 16:23 |
| *** sgotliv_ has quit IRC | 16:26 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: PEP8(E265): block comment should start with '# ' https://review.openstack.org/145821 | 16:26 |
| *** ebalduf has quit IRC | 16:29 | |
| *** Mandell has joined #openstack-cinder | 16:29 | |
| *** EduardBiceriMate has joined #openstack-cinder | 16:30 | |
| *** jkraj has quit IRC | 16:30 | |
| eharney | e0ne: have you seen anything indicating why upgrading helped? | 16:31 |
| e0ne | eharney: a very similar bug in debian https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=659762 | 16:32 |
| *** Mandell has quit IRC | 16:32 | |
| *** Mandell has joined #openstack-cinder | 16:33 | |
| *** chlong has quit IRC | 16:34 | |
| *** ebalduf has joined #openstack-cinder | 16:34 | |
| *** Mandell has quit IRC | 16:38 | |
| *** sgotliv_ has joined #openstack-cinder | 16:39 | |
| *** nellysmitt has quit IRC | 16:44 | |
| openstackgerrit | Yuriy Nesenenko proposed openstack/cinder: Fixed comments style according to the Hacking Rules https://review.openstack.org/145821 | 16:48 |
| *** lpetrut has joined #openstack-cinder | 16:48 | |
| *** lpabon_ has quit IRC | 16:51 | |
| *** jecarey has quit IRC | 16:51 | |
| *** lpetrut has quit IRC | 16:53 | |
| openstackgerrit | Navneet Singh proposed openstack/cinder: Implement Cinder Manage/Unmanage APIs in NFS driver. https://review.openstack.org/138125 | 16:54 |
| nikesh_vedams | Hi We have commited our scst code. https://review.openstack.org/#/c/145778/ To have CI setup,we are following link: https://github.com/j-griffith/sos-ci | 16:55 |
| nikesh_vedams | We have a Ubuntu 14.04 server VM and we will integrate SCST changes in this VM. Now how can we use this VM for CI setup. | 16:55 |
| nikesh_vedams | We are also using this VM for commiting our scst patch | 16:56 |
| nikesh_vedams | We have differnt launchpad account for patch submission and CI account | 16:57 |
| *** TobiasE1 has joined #openstack-cinder | 16:59 | |
| nikesh_vedams | But we want to know how to setup our VM with https://github.com/j-griffith/sos-ci CI for testing our https://review.openstack.org/#/c/145778 patch | 17:00 |
| *** patrickeast has joined #openstack-cinder | 17:00 | |
| nikesh_vedams | please guide us | 17:00 |
| *** _cjones_ has joined #openstack-cinder | 17:00 | |
| *** jecarey has joined #openstack-cinder | 17:00 | |
| *** TobiasE has quit IRC | 17:01 | |
| *** xyang has joined #openstack-cinder | 17:03 | |
| *** e0ne has quit IRC | 17:03 | |
| *** timcl has quit IRC | 17:04 | |
| *** andymaier has joined #openstack-cinder | 17:05 | |
| *** esker has joined #openstack-cinder | 17:07 | |
| tsekiyama | Hi, any suggestion for renaming 'target_portals / target_iqns / target_luns' in iSCSI multipath patch https://review.openstack.org/#/c/134681/ ? thingee told me these need better names when the cinder-spec was approved. | 17:08 |
| jgriffith | hemna: thinking more about detach my thoughts about the detach method signature might have been kinda stupid :) | 17:11 |
| jgriffith | hemna: WRT multi-attach | 17:11 |
| hemna | jgriffith, ah ok. :) I was hoping to circle back on that today | 17:11 |
| jgriffith | hemna: haha... you know me well enough that if you ignore me for a day or two I may change my mind or forget :) | 17:12 |
| hemna | :P | 17:12 |
| hemna | now if I can just get the javelin tests to pass..... | 17:12 |
| openstackgerrit | John Griffith proposed openstack/cinder: Implement clone_image caching on SolidFire https://review.openstack.org/142859 | 17:13 |
| *** EmilienM is now known as EmilienM|afk | 17:15 | |
| jgriffith | winston-d_zZZ: ping... | 17:16 |
| jgriffith | doh! the dreaded zZZ | 17:16 |
| openstackgerrit | Patrick East proposed openstack/cinder: Add support for manage/unmanage volume commands to PureISCSIDriver https://review.openstack.org/143726 | 17:17 |
| *** nshaikh has quit IRC | 17:18 | |
| thingee | jgriffith: did you see all the v1 removal patches? :) | 17:19 |
| jgriffith | thingee: no... | 17:19 |
| thingee | https://review.openstack.org/#/c/145703/ | 17:20 |
| thingee | https://review.openstack.org/#/c/145688/ | 17:20 |
| jgriffith | thingee: got'em | 17:20 |
| jgriffith | thingee: only thing is do we really want to "remove" them? | 17:20 |
| thingee | I thought that was the plan for Kilo. | 17:21 |
| jgriffith | thingee: and part of the only reason I ask is that although we had it enabled in Grizzly, it wasn't default until Juno | 17:21 |
| jgriffith | thingee: it was... | 17:21 |
| thingee | jgriffith: https://github.com/openstack/cinder/blob/master/cinder/api/__init__.py#L31 | 17:21 |
| jgriffith | I'm just making sure we're thinking it through and not shooting ourselves in the foot | 17:21 |
| jgriffith | or pissin a bunch of people off :) | 17:21 |
| thingee | jgriffith: I agree, but we gotta rip the bandaid at some point | 17:22 |
| jgriffith | thingee: yeah... and your log message is pretty clear :) | 17:22 |
| thingee | unfortunately openstackclient, heat, tripleo, etc need to be updated to use the cinderclient itself instead of assuming volumev2 is the service type. so that discovery works | 17:22 |
| jgriffith | ugh | 17:23 |
| jgriffith | thingee: I'm happy to help on some of those.... | 17:23 |
| jgriffith | or at least openstackclient | 17:23 |
| thingee | jgriffith: https://github.com/Thingee/python-openstackclient/commit/4f32c6755346ab5068533bd9c5963ff60ab8e6a6 | 17:23 |
| jgriffith | I'm concerne about some other externals (ansible, shade etc) | 17:23 |
| jgriffith | thingee: oh fine.. take the one that I was interested in helping with :) | 17:24 |
| jgriffith | thingee: I want nothing to do with OOO | 17:24 |
| *** andymaier has quit IRC | 17:24 | |
| jgriffith | thingee: that being said, I'll help out on anything you like, just give me orders :) | 17:24 |
| *** timcl has joined #openstack-cinder | 17:24 | |
| thingee | jgriffith: https://github.com/openstack/heat/blob/master/heat/engine/clients/os/cinder.py#L39 | 17:24 |
| mdenny | \\csim-cloud.rose.hp.com\CloudGroupShare\testing\perf\logs | 17:25 |
| jgriffith | thingee: oh... so did you do all of them already? | 17:25 |
| thingee | jgriffith: nah, that's master of heat..was pointing out the use of volumev2 | 17:25 |
| mdenny | wrong channel | 17:26 |
| *** nkrinner has quit IRC | 17:26 | |
| thingee | jgriffith, jamielennox|away: so I noticed a problem with https://review.openstack.org/#/c/145613/. | 17:27 |
| nikesh_vedams | jgriffith:hi | 17:27 |
| jgriffith | nikesh_vedams: hello there | 17:27 |
| jgriffith | thingee: yeah? | 17:28 |
| thingee | jamielennox|away: besides the problems you pointed out | 17:28 |
| nikesh_vedams | jgriffith:uploaded patch,need some help :) | 17:28 |
| nikesh_vedams | in CI | 17:28 |
| jgriffith | nikesh_vedams: looking | 17:29 |
| thingee | jgriffith: so remember these https://github.com/openstack/python-cinderclient/blob/master/cinderclient/v2/shell.py#L190 ? | 17:29 |
| jgriffith | nikesh_vedams: pep8 and pylint, no big deal | 17:29 |
| jgriffith | thingee: oh | 17:30 |
| nikesh_vedams | jgriffith: we want to know how to setup our VM with https://github.com/j-griffith/sos-ci CI for testing our https://review.openstack.org/#/c/145778 patch | 17:30 |
| jgriffith | thingee: that's problematic | 17:30 |
| jgriffith | nikesh_vedams: so sos-ci is meant to interact with an OpenStack cloud... run devstack and Temest on Instances in that Cloud | 17:31 |
| *** jpich has quit IRC | 17:31 | |
| thingee | jgriffith: these are making my life harder for version discovery with backwards compat. If I have x.x.x.x:8776 as the publicURL and service_type 'volume'...I'll get endpoint not found if I have OS_VOLUME_API_VERSION=1 set. | 17:31 |
| jgriffith | You need to deploy OpenStack and point sos-ci at that | 17:31 |
| thingee | jgriffith: because this will set volumev2 for the service_type https://github.com/openstack/python-cinderclient/blob/master/cinderclient/shell.py#L578 | 17:32 |
| jgriffith | thingee: can we do a mapping trick? | 17:32 |
| jgriffith | thingee: because we can't change the client | 17:32 |
| jgriffith | thingee: I mean, we can't count on the client not being "older" | 17:32 |
| nikesh_vedams | jgriffith: how can we create this Openstack cloud,we have a VM with ubuntu 14.04 server and also we have a ESXi Host | 17:34 |
| thingee | jgriffith: well so first question, we need to decide if it's worth making the cinderclient backwards compat. so either fix up https://review.openstack.org/#/c/145613/ to do some hack of if endpoint not found is raised, we fall back to service_type 'volume' | 17:34 |
| thingee | jgriffith: or | 17:34 |
| thingee | we assume in a major cinderclient release that v1 is gone https://review.openstack.org/#/c/145688/ | 17:34 |
| jgriffith | nikesh_vedams: if you don't know how to deploy OpenStack yet you've got a very very long way to go I'm afraid | 17:34 |
| thingee | and that the service_catalog needs to be updated before upgrading to kilo | 17:34 |
| thingee | service type has to be volume and endpoint has to be x.x.x.x:8776 | 17:35 |
| jgriffith | thingee: but you can't assume people have upgraded the client... I mean you can't break their existing clients IMHO | 17:35 |
| jgriffith | hmmm.... | 17:35 |
| thingee | jgriffith: that's true. | 17:35 |
| jgriffith | thingee: I mean moving everything to default to v2 is def good, and even "hiding" v1 a little | 17:36 |
| nikesh_vedams | jgriffith:so i know how to deploy it by http://docs.openstack.org/juno/install-guide/install/apt/content/ | 17:36 |
| jgriffith | nikesh_vedams: you need to be looking at devstack IMHO | 17:37 |
| *** dustins_ has joined #openstack-cinder | 17:39 | |
| nikesh_vedams | jgriffith: any link will be greatly helpful :) | 17:39 |
| thingee | nikesh_vedams: http://devstack.org | 17:39 |
| *** jwcroppe has quit IRC | 17:40 | |
| *** jwcroppe has joined #openstack-cinder | 17:41 | |
| *** dustins has quit IRC | 17:41 | |
| thingee | jgriffith: if you have an outdated client, I guess we should make it so on the server side, you can service types volume and volumev2 point to x.x.x.x:8776? | 17:41 |
| thingee | jgriffith: I actually haven't thought about that before, but may work...would have to test it | 17:42 |
| *** jistr has quit IRC | 17:43 | |
| *** Masahiro has joined #openstack-cinder | 17:43 | |
| *** jwcroppe has quit IRC | 17:45 | |
| thingee | jgriffith: what a mess :) | 17:46 |
| jgriffith | thingee: sigh | 17:46 |
| thingee | I couldn't sleep for two hours last night because of this. :) | 17:47 |
| jgriffith | thingee: maybe this is why nobody else has really "removed" and API version | 17:47 |
| nikesh_vedams | jgriffith thingee: sorry for silly question but what is IMHO | 17:47 |
| jgriffith | nikesh_vedams: not silly... "IMHO = In My Humble Opinion" | 17:47 |
| *** Masahiro has quit IRC | 17:48 | |
| thingee | jgriffith is very humble with his opinions :) | 17:48 |
| thingee | hey so austin, bbq all week? | 17:48 |
| jgriffith | thingee: kinda of funny that I use that term huh? | 17:48 |
| jgriffith | LOL | 17:48 |
| jgriffith | thingee: I'm a vegetarian | 17:48 |
| jgriffith | NOT! | 17:48 |
| *** rushiagr_away is now known as rushiagr | 17:48 | |
| thingee | jgriffith: ain't foolin' me | 17:48 |
| jgriffith | :) | 17:49 |
| *** sgotliv_ has quit IRC | 17:49 | |
| jgriffith | thingee: so I'm looking at some things on the allocated capacity issues vishy mentioned yesterday | 17:49 |
| jgriffith | thingee: it's kind of a mess | 17:49 |
| jgriffith | thingee: I'm working through some things now, I'll get a bug posted and ping you when it's up | 17:50 |
| *** fandi has quit IRC | 17:50 | |
| nikesh_vedams | jgriifith thingee : hmm then we can give Openstack colud to sos-ci by running devstack right | 17:51 |
| jgriffith | nikesh_vedams: right | 17:51 |
| thingee | jgriffith: for the creates, is it enough to notify the scheduler if we get an exception from the driver? | 17:51 |
| jgriffith | thingee: that's what I'm messing with right now | 17:51 |
| jgriffith | thingee: it "should" | 17:52 |
| thingee | jgriffith: Ok, wasn't sure if it was as simple as that. I tried bugging winston-d_zZZ about it last night, but I never heard from him | 17:52 |
| nikesh_vedams | jgriiffith thingee : thanks | 17:52 |
| thingee | nikesh_vedams: np | 17:52 |
| *** lpetrut has joined #openstack-cinder | 17:52 | |
| openstackgerrit | Thang Pham proposed openstack/cinder-specs: Cinder objects https://review.openstack.org/130044 | 17:52 |
| jgriffith | well, so it's kinda funky... esp with pools in the mix | 17:52 |
| thingee | jgriffith: yeah my head was starting to hurt when I saw the pool docstring :) | 17:52 |
| jgriffith | anyway, I'm testing some things out on it. | 17:52 |
| jgriffith | I'd like to move to just pulling from the driver for all of this | 17:53 |
| thingee | I need to write a cheatsheet so I don't have to keep remembering things | 17:53 |
| jgriffith | and don't even try and count ourselves | 17:53 |
| jgriffith | thingee: welcome to my world :) | 17:53 |
| thingee | jgriffith: +1 | 17:53 |
| jgriffith | thingee: the problem is now that we have 28+ drivers that's a real pain in my *$$ | 17:53 |
| jgriffith | thingee: and looking at some of them and trying to determine how the hell they work is not fun | 17:54 |
| jgriffith | thingee: so maybe a driver impl, with a fall back | 17:54 |
| jgriffith | anyway.... need to waste another hour of my life in a pointless meeting, then be back to look at it again | 17:54 |
| *** karimb has quit IRC | 17:55 | |
| *** TobiasE1 has quit IRC | 17:57 | |
| rushiagr | jgriffith: that's a good time to check out all social networking sites on your phone ;) | 17:58 |
| rushiagr | I'm not back from holidays yet, so I think I should shut my lappy down and enjoy instead :) | 18:00 |
| rushiagr | good night :) | 18:00 |
| *** jordanP has quit IRC | 18:01 | |
| *** fandi has joined #openstack-cinder | 18:02 | |
| *** jwcroppe has joined #openstack-cinder | 18:04 | |
| nikesh_vedams | jgriffith thingee :Like you said that sos-ci will run devstack and Temest on Instances in that Cloud,so will it create instance itself or we have to create instance manually? | 18:04 |
| *** leeantho has joined #openstack-cinder | 18:04 | |
| *** rushiagr is now known as rushiagr_away | 18:06 | |
| thingee | nikesh_vedams: it will create the instance | 18:06 |
| nikesh_vedams | jgriffith thingee : ok but i think we have to manually install scst packages in those VMs | 18:08 |
| nikesh_vedams | jgriffith thingee : without scst packages tempest will fail | 18:09 |
| *** dannywilson has joined #openstack-cinder | 18:09 | |
| nikesh_vedams | jgriffith thingee : i mean if sos-ci create VM and deploy devstack in VM and try to run tempest,it will fail without scst packages | 18:10 |
| thingee | nikesh_vedams: then you should modify sos-ci for your case. When it installs devstack it has a place to install additional packages https://github.com/j-griffith/sos-ci/blob/master/sos-ci/ansible/tasks/install_devstack.yml#L4 | 18:12 |
| thingee | assuming they are available via apt-get | 18:12 |
| openstackgerrit | Mudassir Latif proposed openstack/cinder: Add replication and PG setup methods on PureISCSIDriver https://review.openstack.org/145090 | 18:12 |
| *** EmilienM|afk is now known as EmilienM | 18:14 | |
| *** annashen has joined #openstack-cinder | 18:18 | |
| nikesh_vedams | jgriffith thingee : ohhh,we are getting scst packages through wget and then we are doing "make and make install" for installing scst packages and some other commands we are running for scst | 18:19 |
| nikesh_vedams | running | 18:20 |
| nikesh_vedams | and also we are editing /etc/rootwrap.d/ | 18:20 |
| thingee | nikesh_vedams: Ok, so you'll need modify that file so ansible does that | 18:20 |
| *** MIDENN_ has quit IRC | 18:23 | |
| *** mdenny has quit IRC | 18:23 | |
| *** e0ne has joined #openstack-cinder | 18:24 | |
| *** mdenny has joined #openstack-cinder | 18:25 | |
| thingee | jgriffith: ok whew, this will work if we backwards compat http://paste.openstack.org/show/156006/ | 18:25 |
| *** leeantho has quit IRC | 18:25 | |
| *** mdenny has quit IRC | 18:25 | |
| *** bswartz has quit IRC | 18:30 | |
| *** mdenny has joined #openstack-cinder | 18:33 | |
| *** e0ne has quit IRC | 18:34 | |
| *** e0ne has joined #openstack-cinder | 18:37 | |
| *** junhongl has quit IRC | 18:38 | |
| *** lpabon has quit IRC | 18:39 | |
| *** junhongl has joined #openstack-cinder | 18:39 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/cinder: Updated from global requirements https://review.openstack.org/145661 | 18:42 |
| *** sgotliv has joined #openstack-cinder | 18:43 | |
| *** nellysmitt has joined #openstack-cinder | 18:45 | |
| *** Mandell has joined #openstack-cinder | 18:47 | |
| *** _cjones_ has quit IRC | 18:48 | |
| thingee | jgriffith: now to deal with volumev2 being removed =/ | 18:49 |
| *** _cjones_ has joined #openstack-cinder | 18:49 | |
| *** nellysmitt has quit IRC | 18:50 | |
| *** Mandell_ has joined #openstack-cinder | 18:50 | |
| *** Mandell has quit IRC | 18:51 | |
| *** jecarey has quit IRC | 18:59 | |
| nikesh_vedams | thingee : one more question ,how sos-ci will take my patch and CI account information :) | 19:01 |
| thingee | nikesh_vedams: environment variables or hard coded https://github.com/j-griffith/sos-ci/blob/master/sos-ci/os_ci.py#L17 | 19:02 |
| thingee | nikesh_vedams: the patch comes from the event queue that is being tested with | 19:04 |
| *** jecarey has joined #openstack-cinder | 19:05 | |
| *** adrien1 has joined #openstack-cinder | 19:05 | |
| *** adrien1 is now known as adrienverge | 19:05 | |
| adrienverge | Hi everyone | 19:05 |
| adrienverge | A new feature got merged into Cinder and I would like to have it documented in http://docs.openstack.org/trunk/config-reference/content/ | 19:06 |
| adrienverge | How can I contribute? Where's the source of this config-reference? | 19:07 |
| *** mriedem has quit IRC | 19:07 | |
| nikesh_vedams | thingee: i want it to run only patches for this "https://review.openstack.org/#/c/145778/" | 19:08 |
| thingee | nikesh_vedams: If you want to do an individual patch, I would recommend just creating a vm manually with devstack, but in the local.conf specify the CINDER_REPO and CINDER_BRANCH as explained here http://docs.openstack.org/developer/devstack/stackrc.html?highlight=_repo | 19:10 |
| thingee | nikesh_vedams: no need for sos-ci if that's the case | 19:11 |
| thingee | nikesh_vedams: sos-ci is for monitoring upstream events and automating the testing | 19:11 |
| *** alexpilotti has quit IRC | 19:14 | |
| *** adrienverge has quit IRC | 19:16 | |
| nikesh_vedams | thingee: but by this way how coud we provide resut of tempest log to gerrit | 19:18 |
| *** ebalduf has quit IRC | 19:18 | |
| nikesh_vedams | thingee: we must need some CI to provide results to gerrit right | 19:18 |
| nikesh_vedams | and so i was wondering how sos-ci wil take only our patch and provide resluts to gerrit | 19:20 |
| *** mriedem has joined #openstack-cinder | 19:22 | |
| *** annashen has quit IRC | 19:23 | |
| *** annashen has joined #openstack-cinder | 19:24 | |
| *** EduardBiceriMate has quit IRC | 19:24 | |
| timcl | ameade: to add your phone go here - http://hr.netapp.com/how-do-i/ | 19:25 |
| timcl | select Change - Phone #s | 19:26 |
| timcl | fill out and submit | 19:26 |
| timcl | oops wrong window | 19:26 |
| ameade_ | hehe | 19:27 |
| *** ameade_ is now known as ameade | 19:29 | |
| *** Masahiro has joined #openstack-cinder | 19:31 | |
| thingee | nikesh_vedams: if you have the ci setup, and reupload the patch, the idea is the ci would pull in the patch. | 19:32 |
| *** _cjones_ has quit IRC | 19:33 | |
| *** jecarey has quit IRC | 19:33 | |
| *** jecarey has joined #openstack-cinder | 19:33 | |
| *** _cjones_ has joined #openstack-cinder | 19:34 | |
| *** Longgeek has quit IRC | 19:34 | |
| *** Masahiro has quit IRC | 19:35 | |
| *** Longgeek has joined #openstack-cinder | 19:36 | |
| *** dalgaaf has quit IRC | 19:37 | |
| nikesh_vedams | thingee: ok but i have fear that it will take other patches too along with my patch so there should be some restrictions | 19:38 |
| *** annashen has quit IRC | 19:39 | |
| thingee | nikesh_vedams: that's the idea of a CI. If that's not what you want, then you don't want continous integration. | 19:40 |
| thingee | nikesh_vedams: if you want to do a one time test, start a vm manually, install whatevwer, run tempest, record the logs, upload it somewhere, post it in comments of the review. | 19:40 |
| *** Longgeek has quit IRC | 19:41 | |
| *** e0ne is now known as e0ne_ | 19:41 | |
| nikesh_vedams | thingee:For our patch what is good according to community https://review.openstack.org/#/c/145778/ | 19:42 |
| nikesh_vedams | thingee: having CI or manuall logs | 19:43 |
| thingee | nikesh_vedams: definitely having a ci. If you test manually now, it may work now, but might not work later. | 19:44 |
| thingee | nikesh_vedams: a CI would report if any proposed patch would break scst support | 19:45 |
| thingee | because you are continuously testing | 19:45 |
| jgriffith | vishy: so it seems to me that to start, we should increment allocated cap regardless of the result from the driver on create | 19:47 |
| jgriffith | vishy: the volume's going to be there and allocated regardless | 19:47 |
| *** timcl has quit IRC | 19:47 | |
| jgriffith | by "there" I mean in Cinder's DB | 19:47 |
| thingee | jgriffith: we don't know where the driver failed though | 19:48 |
| jgriffith | thingee: doesn't matter | 19:48 |
| jgriffith | thingee: it's "allocated" | 19:48 |
| jgriffith | thingee: in Cinder | 19:48 |
| *** e0ne_ is now known as e0ne | 19:48 | |
| jgriffith | even if it's not used on the backend a request was made to allocate the space | 19:48 |
| nikesh_vedams | jgriffith thingee : thanks for giving your valueable time to us :) | 19:48 |
| thingee | jgriffith: I get that, but I guess it doesn't seem accurate. not saying it's any better than now. | 19:49 |
| jgriffith | thingee: but it is accurate | 19:49 |
| openstackgerrit | Steven Kaufer proposed openstack/python-cinderclient: WIP Adopt CLI Sorting Argument Guidelines https://review.openstack.org/145913 | 19:49 |
| jgriffith | thingee: as opposed ot now which is NOT at all accurate | 19:49 |
| jgriffith | thingee: so look at it like this.... | 19:49 |
| jgriffith | you create a 1G volume.... acap = 1G | 19:50 |
| jgriffith | you create another 1G vol, but it fails... | 19:50 |
| jgriffith | You STILL have 2Gig of storage allocated from Cinder's perspective | 19:50 |
| jgriffith | until you delete the volume that's in error state | 19:50 |
| jgriffith | that being said.... | 19:51 |
| jgriffith | the next step I'd like is to say; "hey, get this info via a call to the driver" | 19:51 |
| jgriffith | and then, if that's not implemented fall back and use the local tracking | 19:51 |
| thingee | but cinder is going to think 2 gigs is allocated, when it could've been a network issue between the cinder manager and the backend. The backend never knew about it. | 19:51 |
| thingee | unless I'm still missing something here. | 19:51 |
| jgriffith | thingee: but it IS allocated!!! | 19:51 |
| jgriffith | thingee: there's a volume of 1G in the database | 19:51 |
| jgriffith | thingee: you do a volume list and there it is | 19:52 |
| jgriffith | that's allocated, | 19:52 |
| jgriffith | maybe not "used" but it's been allocated | 19:52 |
| jgriffith | thingee: see what I mean? | 19:52 |
| openstackgerrit | Steven Kaufer proposed openstack/python-cinderclient: WIP Adopt CLI Sorting Argument Guidelines https://review.openstack.org/145913 | 19:52 |
| thingee | Yeah i understand now. I'm not sure I have a good understanding of what allocated means to the scheduler here then. | 19:52 |
| jgriffith | thingee: so for scheduler that's how the weigher works and the capacity filter etc | 19:53 |
| *** _cjones_ has quit IRC | 19:53 | |
| jgriffith | thingee: the thing is there's a difference between allocation and utilization | 19:53 |
| jgriffith | and as long as we're clear and consistent on that I think it's "ok" | 19:53 |
| thingee | because I was under the impression is was as best as we can guess is allocated from the backend. Unless we get to the expected state, it's not counted. | 19:54 |
| thingee | not disagreeing, just talking through here. | 19:54 |
| jgriffith | thingee: oh sure... yeah, talk is good :) | 19:54 |
| jgriffith | thingee: but I'm not sure I understand your statement there? | 19:54 |
| jgriffith | thingee: so this all kinda comes back to what we've grappled with for a while in terms of consistent definitions and reporting | 19:55 |
| *** ebalduf has joined #openstack-cinder | 19:56 | |
| jgriffith | honestly I don't think it matters if we define blue as red and green as purple.... as long as it's consistent | 19:56 |
| *** thangp has quit IRC | 19:56 | |
| * jgriffith doesn't know if that analogy helps or makes things worse :) | 19:56 | |
| thingee | so with what you were saying, allocated makes sense from cinder's perspective and is more accurate. I agree with that. However, it's not any better of being accurate of the backend, and that's not going to happen until we actually get reporting from the backend itself. | 19:56 |
| thingee | am I accurate with that? :P | 19:56 |
| jgriffith | thingee: well it's way better actually | 19:56 |
| jgriffith | thingee: cuz right now if you have 5 volumes, create one and it goes to error you don't increment | 19:57 |
| jgriffith | BUT... when you delete that volume it does decrement :( | 19:57 |
| jgriffith | thingee: and it's nothing new in the storage industry to differentiate between allocated and used space | 19:57 |
| thingee | sure | 19:58 |
| thingee | I think I got it now | 19:58 |
| jgriffith | and since Cinder is the application here we say "yes, you allocated it. Even if you didn't actually reserve any blocks on your device, you did in fact allocate N blocks from backend-foo" | 19:58 |
| thingee | yeah seems reasonable to me. would be curious vishy's thoughts. | 19:58 |
| jgriffith | thingee: vishy it's def just an interim improvement | 19:59 |
| openstackgerrit | Patrick East proposed openstack/cinder: Add support for manage/unmanage volume commands to PureISCSIDriver https://review.openstack.org/143726 | 19:59 |
| *** jecarey has quit IRC | 19:59 | |
| jgriffith | thingee: there's some other weird things in here too, but that's a start anyway | 20:00 |
| *** jecarey has joined #openstack-cinder | 20:01 | |
| jgriffith | thingee: hmmm.... so tricky part is if we do a revert on failure to create | 20:08 |
| openstackgerrit | Navneet Singh proposed openstack/cinder: Implement Cinder Manage/Unmanage APIs in NFS driver. https://review.openstack.org/138125 | 20:08 |
| jgriffith | thingee: or I should say if "taskflow" does a revert | 20:08 |
| thingee | probably taskflow, in case retries | 20:09 |
| jgriffith | thingee: what I mean is we track in the manager, we don't know what taskflow is doing higher up :( | 20:09 |
| jgriffith | thingee: in most cases you just get a volume in error state, but in some it actually reverts it | 20:10 |
| jgriffith | so the volume gets nuked out of db, BUT your allocated cap isn't adjusted | 20:10 |
| jgriffith | and the hooks into manager's cap reporting aren't exactly "readily" available | 20:10 |
| jgriffith | hmm... guess I'll look at this some more | 20:11 |
| *** fandi has quit IRC | 20:13 | |
| thingee | jgriffith: wait, I thought for allocated we were going to just go by what's known in the cinder db, instead of incrementing and decrementing. Would it not matter then? | 20:14 |
| jgriffith | thingee: yes | 20:20 |
| jgriffith | thingee: but... taskflow calls manager | 20:20 |
| jgriffith | taskflow creates vol in db.... calls manager | 20:20 |
| jgriffith | manager does allocation stuff | 20:20 |
| jgriffith | returns to taskflow | 20:20 |
| *** annashen has joined #openstack-cinder | 20:20 | |
| jgriffith | problem is that if something goes wrong, taskflow may or may not delete the volume from db | 20:20 |
| jgriffith | the retries is a non-issue | 20:21 |
| jgriffith | but if it reverts we've allocated and we're screwed | 20:21 |
| jgriffith | but I think I know how to work that | 20:21 |
| jgriffith | so never mind :) | 20:21 |
| thingee | jgriffith: I think I might still be missing. so if you have 1G volume, alloc=1G....then you do another 1G create, fails in taskflow, the volume is not deleted by taskflow..so you have 2 x 1G volumes know to cinder. One is error the other available. If we remove out the increment/decrement code and just purely go off what's know in the cinder db. it will still be alloc 2G. Now if taskflow was to | 20:24 |
| thingee | delete volume, we would still go off what's known in the cinder db, and be at 1G. | 20:24 |
| *** leeantho has joined #openstack-cinder | 20:24 | |
| jgriffith | thingee: so that scenario is good... yes | 20:24 |
| thingee | those are scenarios | 20:24 |
| thingee | two* | 20:24 |
| thingee | I'm trying to understand what the manager has to do with any of that if the scheduler is just asking the cinder db directly. there is no incrementing or decrementing | 20:25 |
| jgriffith | thingee: ahh.... so you're saying just skip the update on create/delete altogether maybe and just do the periodic task of checking db and counting refs? | 20:25 |
| thingee | yes | 20:25 |
| jgriffith | thingee: so the problem is you do a create of 10 volumes at once | 20:26 |
| jgriffith | thingee: your capacity info isn't updated until the next periodic | 20:26 |
| jgriffith | bad things happen | 20:26 |
| thingee | got it | 20:26 |
| thingee | jgriffith: the other problem to is if reschedule fails in taskflow. | 20:30 |
| thingee | just glancing at the code, might be a concern | 20:30 |
| *** fifieldt_ has quit IRC | 20:32 | |
| thingee | jgriffith: actually, I can't find anywhere in taskflow manager create volume code where the volume is explicitly deleted from the db | 20:33 |
| thingee | just the common.error_out_volume() | 20:33 |
| *** fifieldt_ has joined #openstack-cinder | 20:33 | |
| *** Mandell_ has quit IRC | 20:34 | |
| e0ne | thingee: just 2c from my side: afaik, our flow manager doesn't support anything except creating | 20:37 |
| *** harlowja_away is now known as harlowja | 20:37 | |
| e0ne | hope, i'm wrong | 20:37 |
| thingee | e0ne: that's correct | 20:37 |
| e0ne | :( | 20:37 |
| thingee | e0ne: eharney signed up to delete volume next | 20:38 |
| e0ne | thingee, eharney: great news | 20:38 |
| thingee | e0ne: as discussed in paris, we wanted something that would be more useful to use taskflow to determine if wanted to keep using taskflow at all | 20:38 |
| e0ne | thingee: i'm looking on it and starting of state machine impl for a last few days | 20:39 |
| *** kaufer has quit IRC | 20:39 | |
| *** jecarey has quit IRC | 20:40 | |
| *** jecarey has joined #openstack-cinder | 20:41 | |
| *** e0ne has quit IRC | 20:41 | |
| *** jecarey_ has joined #openstack-cinder | 20:42 | |
| *** bswartz has joined #openstack-cinder | 20:43 | |
| *** lpetrut has quit IRC | 20:43 | |
| *** kvidvans has quit IRC | 20:44 | |
| *** jecarey has quit IRC | 20:45 | |
| *** e0ne has joined #openstack-cinder | 20:46 | |
| *** nellysmitt has joined #openstack-cinder | 20:46 | |
| *** nellysmitt has quit IRC | 20:51 | |
| *** _cjones_ has joined #openstack-cinder | 20:53 | |
| *** david-lyle has joined #openstack-cinder | 20:54 | |
| *** lpabon has joined #openstack-cinder | 20:58 | |
| boris-42 | thingee: jgriffith hey guys | 20:58 |
| boris-42 | thingee: jgriffith can we get this in https://review.openstack.org/#/c/145595/ | 20:58 |
| boris-42 | seems like I put proper numbers now | 20:58 |
| *** _cjones_ has quit IRC | 20:59 | |
| *** _cjones_ has joined #openstack-cinder | 20:59 | |
| *** Mandell has joined #openstack-cinder | 20:59 | |
| jgriffith | boris-42: yeah, once we get Jenkins coerced | 21:02 |
| boris-42 | jgriffith: thanks | 21:03 |
| *** markvoelker has joined #openstack-cinder | 21:04 | |
| *** kaufer has joined #openstack-cinder | 21:06 | |
| *** e0ne has quit IRC | 21:09 | |
| rushil | jgriffith, thingee, xyang, avishay, jungleboyj: Can you look at https://review.openstack.org/#/c/143431/ ? I have included the reasoning for the change in both the commit message and the bug. Thanks | 21:10 |
| *** e0ne has joined #openstack-cinder | 21:12 | |
| *** mriedem has quit IRC | 21:13 | |
| *** mriedem has joined #openstack-cinder | 21:15 | |
| thingee | boris-42: sure. e0ne left you a comment tho | 21:18 |
| *** lpetrut has joined #openstack-cinder | 21:18 | |
| *** patrickeast has quit IRC | 21:18 | |
| thingee | jgriffith: did you see my earlier commments about not finding explicit deletes in the taskflow manager create volume code? | 21:19 |
| thingee | jgriffith: all I found was common.error_out_volume() | 21:19 |
| *** Masahiro has joined #openstack-cinder | 21:20 | |
| jgriffith | thingee: https://github.com/openstack/cinder/blob/master/cinder/volume/flows/api/create_volume.py#L522 | 21:20 |
| thingee | does the manager talk to the api taskflow code? I thought it was just the manager taskflow code | 21:21 |
| thingee | jgriffith: ^ | 21:21 |
| jgriffith | thingee: you tell me, you and DuncanT keep saying how easy taskflow is :) | 21:21 |
| jgriffith | thingee: or at least DuncanT did, not sure if you did or not :) | 21:22 |
| jgriffith | thingee: I'm trying to force that whitelist to trace through it | 21:22 |
| e0ne | :) | 21:22 |
| jgriffith | maybe harlowja is around? | 21:23 |
| thingee | jgriffith: my impression was api/taskflow was just for api code to talk to. | 21:23 |
| jgriffith | thingee: that would make sense | 21:23 |
| harlowja | jgriffith what u saying, lol | 21:23 |
| jgriffith | harlowja: long time no hear | 21:24 |
| harlowja | ya, happy new year | 21:24 |
| harlowja | i got a case of poison oak for xmas | 21:24 |
| harlowja | fun fun | 21:24 |
| harlowja | lol | 21:24 |
| thingee | harlowja: is it safe to assume volume/flows/api/create_volume.py is only access from the cinder api code, not the manager? | 21:24 |
| *** jamielennox|away is now known as jamielennox | 21:24 | |
| jgriffith | harlowja: revert routine on taskflow create volume | 21:24 |
| *** patrickeast has joined #openstack-cinder | 21:24 | |
| jgriffith | harlowja: does taskflow ever go back and delete the db ref on a failure? | 21:24 |
| harlowja | thingee afaik ya, tried to keep them under a specific folder (api -> api) | 21:24 |
| *** Masahiro has quit IRC | 21:25 | |
| thingee | harlowja: from what I gather, when the manager talks to taskflow, taskflow will only set the volume to error state. A volume is only deleted if something fails from the cinder api side. | 21:26 |
| thingee | before getting to the manager/scheduler, etc. | 21:26 |
| harlowja | ya, that should be the expected case | 21:26 |
| jgriffith | harlowja: thingee I'm not sure what you folks are referring to | 21:27 |
| thingee | jgriffith: so the cinder api code, talk to volume/flows/api/create_volume.py. the manager should be just talking to volume/flows/manager/create_volume.py | 21:28 |
| jgriffith | thingee: ummm... nope | 21:28 |
| jgriffith | thingee: :) | 21:29 |
| harlowja | https://github.com/openstack/cinder/blob/master/cinder/volume/flows/manager/create_volume.py#L147 | 21:29 |
| jgriffith | create volume from cinder api goes to flows/api/create_volume | 21:29 |
| jgriffith | that's where the db entry is created | 21:29 |
| jgriffith | :) | 21:29 |
| harlowja | ya | 21:30 |
| jgriffith | which is the link I sent | 21:30 |
| jgriffith | and it's also where the rever method is that does a db.volume_destroy | 21:30 |
| jgriffith | simple :) | 21:30 |
| harlowja | ok maybe i'm confused to then, ha, no api workflow code should be used in other workflows | 21:31 |
| harlowja | i think thats the jist of the question | 21:31 |
| jgriffith | harlowja: thingee I could be confusing everybody.... | 21:31 |
| jgriffith | harlowja: thingee let's start over :) | 21:31 |
| harlowja | possibly, ha | 21:31 |
| jgriffith | harlowja: thingee my question was..... | 21:31 |
| jgriffith | harlowja: thingee under what circumstances can the revert method get called (here: https://github.com/openstack/cinder/blob/master/cinder/volume/flows/api/create_volume.py#L511) | 21:32 |
| harlowja | when a later task fails | 21:32 |
| *** lpabon has quit IRC | 21:32 | |
| jgriffith | harlowja: thingee and yes, volume create calls go through here | 21:32 |
| harlowja | which is the quota commit task, VolumeCastTask | 21:32 |
| jgriffith | harlowja: ahh... so create was fine (or not) but quota update fails | 21:33 |
| jgriffith | so then we revert? | 21:33 |
| harlowja | seems like it | 21:33 |
| harlowja | that would be one case of that, ya | 21:33 |
| harlowja | or if casting fails, it should get there also | 21:34 |
| harlowja | but at that point someone set 'optional_args['is_quota_committed']' | 21:34 |
| harlowja | bb | 21:34 |
| *** jgravel has joined #openstack-cinder | 21:37 | |
| *** eharney has quit IRC | 21:39 | |
| *** Mandell_ has joined #openstack-cinder | 21:45 | |
| *** Mandell has quit IRC | 21:46 | |
| *** Tross has joined #openstack-cinder | 21:47 | |
| *** ryanpetrello has quit IRC | 21:53 | |
| *** e0ne has quit IRC | 21:54 | |
| openstackgerrit | Gloria Gu proposed openstack/python-cinderclient: Update to change name for volume type client https://review.openstack.org/141187 | 21:55 |
| *** e0ne has joined #openstack-cinder | 21:58 | |
| *** dannywilson has quit IRC | 21:59 | |
| *** dannywilson has joined #openstack-cinder | 22:00 | |
| *** patrickeast has quit IRC | 22:01 | |
| *** dannywilson has quit IRC | 22:01 | |
| *** dannywilson has joined #openstack-cinder | 22:01 | |
| *** patrickeast has joined #openstack-cinder | 22:02 | |
| *** dannywilson has quit IRC | 22:02 | |
| *** dannywilson has joined #openstack-cinder | 22:02 | |
| *** fischerw has quit IRC | 22:03 | |
| *** fischerw has joined #openstack-cinder | 22:03 | |
| *** e0ne has quit IRC | 22:05 | |
| *** fischerw has quit IRC | 22:09 | |
| *** xyang has quit IRC | 22:10 | |
| *** xyang has joined #openstack-cinder | 22:11 | |
| gugl | thingee, xyang , this is the change splitted out from the original change to have more discussion regarding whether to allow the volume type name change : https://review.openstack.org/#/c/140906/ whenever you have a change please shed some light. Thank a lot! | 22:14 |
| gugl | thingee, xyang s/change/chance | 22:14 |
| *** fischerw has joined #openstack-cinder | 22:15 | |
| *** EmilienM is now known as EmilienM|afk | 22:17 | |
| *** Mandell_ has quit IRC | 22:22 | |
| *** Yogi1 has quit IRC | 22:22 | |
| *** lcurtis has quit IRC | 22:26 | |
| *** markstur has quit IRC | 22:27 | |
| *** david-lyle has quit IRC | 22:28 | |
| *** mwichmann has quit IRC | 22:29 | |
| *** e0ne has joined #openstack-cinder | 22:34 | |
| *** kaufer has quit IRC | 22:35 | |
| *** jecarey_ has quit IRC | 22:38 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Create Consistency Group from CG Snapshot https://review.openstack.org/145952 | 22:41 |
| *** markstur has joined #openstack-cinder | 22:43 | |
| *** delattec has quit IRC | 22:43 | |
| openstackgerrit | xing-yang proposed openstack/cinder: Create Consistency Group from CG Snapshot https://review.openstack.org/145952 | 22:45 |
| *** Mandell has joined #openstack-cinder | 22:47 | |
| *** nellysmitt has joined #openstack-cinder | 22:47 | |
| *** e0ne has quit IRC | 22:47 | |
| *** mriedem has quit IRC | 22:49 | |
| *** xyang has quit IRC | 22:51 | |
| *** nellysmitt has quit IRC | 22:51 | |
| *** annashen has quit IRC | 22:58 | |
| *** patrickeast has quit IRC | 22:59 | |
| *** dannywilson has quit IRC | 22:59 | |
| *** dannywilson has joined #openstack-cinder | 23:01 | |
| *** patrickeast has joined #openstack-cinder | 23:02 | |
| *** annashen has joined #openstack-cinder | 23:03 | |
| *** btran has quit IRC | 23:05 | |
| *** btran has joined #openstack-cinder | 23:05 | |
| *** Masahiro has joined #openstack-cinder | 23:09 | |
| *** esker has quit IRC | 23:09 | |
| *** Masahiro has quit IRC | 23:13 | |
| *** jwcroppe has quit IRC | 23:15 | |
| *** jwcroppe has joined #openstack-cinder | 23:15 | |
| *** jwcroppe has quit IRC | 23:20 | |
| *** eharney has joined #openstack-cinder | 23:21 | |
| *** eharney_ has joined #openstack-cinder | 23:25 | |
| *** eharney has quit IRC | 23:26 | |
| *** lpetrut has quit IRC | 23:28 | |
| *** dustins_ has quit IRC | 23:28 | |
| *** fischerw has quit IRC | 23:31 | |
| *** jcru has quit IRC | 23:31 | |
| *** asselin has quit IRC | 23:33 | |
| *** asselin has joined #openstack-cinder | 23:38 | |
| *** krtaylor has quit IRC | 23:39 | |
| *** akerr_ has quit IRC | 23:40 | |
| *** erlon_ has quit IRC | 23:40 | |
| openstackgerrit | Nagendra Rao Jaladanki proposed openstack/cinder-specs: Spec for Brocade Zone Driver Virtual Fabrics support https://review.openstack.org/144389 | 23:42 |
| openstackgerrit | John Griffith proposed openstack/cinder: Implement clone_image caching on SolidFire https://review.openstack.org/142859 | 23:46 |
| openstackgerrit | Anthony Lee proposed openstack/cinder: Fixed server name being retained after detach in LeftHand https://review.openstack.org/124145 | 23:48 |
| *** krtaylor has joined #openstack-cinder | 23:52 | |
| *** dannywilson has quit IRC | 23:55 | |
| *** dannywilson has joined #openstack-cinder | 23:56 | |
| *** smoriya has joined #openstack-cinder | 23:59 | |
| *** patrickeast has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!